Hi Reader,

Getting started with AI in finance feels overwhelming.
You try ChatGPT once…, and it gives you a half-useful answer… and then what?

Most AI roadmaps are either way too technical—or just buzzwords with no substance.

So, I built one that actually makes sense.
No fluff. No PhD required.

These are just real steps, from “I have no idea what I’m doing” to “I built an AI that reads invoices.”

But before I present my AI roadmap to you, I want to let you know about an event coming up next week on How to Become an AI CFO.

This event is reserved for CFOs, Fractional CFOs, and finance executives, and the spots are limited, so I encourage you to sign up quickly as the interest is huge and its totally free to join!

What if 1 hour could change how you work forever?

In this free webinar, I’ll show you:

• How AI is already transforming the CFO role—and what you need to do to keep up
• The biggest mistakes finance teams make with AI (and how to avoid falling into the same traps)
• The exact tools top CFOs are using for forecasting, automation, and financial analysis

Bonus: Real-life examples of how AI saved me hours!

​Click here to secure your spot now​

5 Stages of Implementing AI in Finance

From “I don’t know where to start” to “I built an AI that reads invoices.”

Most AI roadmaps are either too technical or too fluffy.
This one is built by finance pros, for finance pros—to get you from Beginner to Master.

Stage 1: Beginner

You’ve heard of ChatGPT. Maybe even tried it once.
But you’re stuck. Worried about confidentiality?

Unsure what it can actually do for finance.

Here’s how to break the ice with AI:

• Today: Write down all the small tasks you do at work. Yes, all of them.
• Tomorrow: Pick one of these tasks and go into ChatGPT, Copilot, or Gemini and ask:

​
“My job is X. I need to do [task description + tool used for it]. Guide me in making me super productive.”

Look at this prompt example:

It might only work 20% of the time.
But out of 10 times you try per day, that’s still 2 wins a day you didn’t have before.

Lesson:
​Stop using AI like Google. Start using it like an intern and a free consultant.

⚠️ Reminder: Don’t enter confidential info if you don’t have an enterprise license. No client names. No sensitive numbers.

Stage 2: Basic

Now you’ve seen the magic—and the misses.
Time to stop guessing and start prompting like a pro.

Here’s the prompt framework I teach in every course and workshop:

• CSI → Context / Specific / Instruction
• + FBI → Format / Blueprint / Identity
  Put them together: CSI + FBI

Want consistent, valuable results? This is the cheat code.
More than 5,000 finance pros have already learned it.

Stage 3: Intermediate

You’re getting results.
But throw a complex problem at AI… and it chokes.

That’s when prompt engineering comes in.

3 techniques to master:

• Chain of Thought → For complex reasoning
• Chunking → For step-by-step processes
• Agent Prompting → For end-to-end financial analysis
  ​

Want a shortcut?

I’ve packaged all of this into:

• A crash course on ChatGPT for Finance
• 10 prompt engineering patterns
• A full Copilot walkthrough

👉 Start here if you want to go from “experimenting” to “executing.”

Stage 4: Advanced

You’re crushing it inside ChatGPT.
But… confidential data? Big files? Automation? Forget it.

Here’s the scary next step: Python.

Why?

Because Python can:

• Crunch numbers
• Merge Excel files
• Process massive datasets
• Create graphics
• Do it all in a secure environment

And no, you don’t need to spend 6 months learning it.

We built a method to learn Python in 4 hours using AI itself.

Over 200 finance pros have done it. They rated the course 4.7/5.

If you want to automate financial analysis without becoming a developer,
👉 Enroll in the next cohort here.

Stage 5: Master

This is the level where AI actually changes how you work.

• You’ve integrated AI into your workflows
• You’ve chosen the right tools for your organization
• You’re building finance-specific agents with low-code platforms like n8n

For example, you can literally use AI to read invoices with image recognition and extract all the information in a spreadsheet.

No coding experience is required.

This is what I did last week and what we teach inside the AI Finance Club (we have a masterclass tomorrow on using n8n to build an AI agent).

We’re building the future of finance—one automation at a time.

👉 Join us in the club now​
​

Best,
Nicolas
​
​PS. The demand for AI CFOs will only grow during the year.

​Start your journey of becoming one by joining my next webinar(hurry up, last time we were at capacity)​